Deep Neural Network-Based Receiver for Next-Generation LEO Satellite Communications

Zhang, Yufeng; Wang, Zhugang; Huang, Yonghui; Ren, Jian; Yin, Yingzeng; Liu, Ying; Pedersen, Gert Frølund; Shen, Ming

doi:10.1109/access.2020.3044321

Cited by 7 publications

(3 citation statements)

References 16 publications

(25 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, DL as a tool for interference mitigation has not been extensively showcased in a satellite-based Internet of Things (S-IoT) scenario. For a satellite receiver, a DL receiver has been shown in [31], however, the work focuses on high throughput transmissions and the non-linear impairments produced by the hardware. While conventional DL works well at detecting signals with non-Gaussian noise, the high power requirements of complex DL models are not ideal for use in systems with limited power availability such as onboard processors in LEO satellites.…”

Section: Related Workmentioning

confidence: 99%

Spiking Neural Networks for Detecting Satellite-Based Internet-of-Things Signals

Dakic¹,

Homssi²,

Walia³

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>The obtained results demonstrate a significantly enhanced detection performance under heavy interference conditions when employing spiking-based and deep learning detection techniques, as opposed to traditional baseline matched-filter methods. Although spiking-based detection approaches exhibit error rates comparable to those of deep learning techniques, they consume considerably less power – several orders of magnitude lower, in fact. Owing to their power efficiency, spiking-based detection networks emerge as the optimal choice for signal detection in resource-limited systems, such as low-Earth orbit satellites.</p>

show abstract

Section: Related Workmentioning

confidence: 99%

Spiking Neural Networks for Detecting Satellite-Based Internet-of-Things Signals

Dakic¹,

Homssi²,

Walia³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The conventional scheme for ML algorithms is to combine all information in a central way and then expose the learning problem in the network. But, considering that massive amount of information is essential to train the DNN scheme [87][88], this increases the computation overhead and latency. Furthermore, with the appearance of a diversity of private providers of minor satellites, it can be forbidden to share the information due to confidentiality or information provider ideas.…”

Section: B Satellite Communicationsmentioning

confidence: 99%

Distributed Learning for 6G–IoT Networks: A Comprehensive Survey

Das¹,

Mudi²,

Rahman³

et al. 2022

Preprint

View full text Add to dashboard Cite

<p>Smart services based on the Internet of Things (IoT) are likely to grow in popularity in the forthcoming years, necessitating the improvement of fifth-generation (5G) cellular networks upgrade of future networks from their present state. Despite the fact that the 5G cellular networks may manage a diversity of IoT services, they may not be able to fully meet the requirements of emerging smart applications due to their limitations that, in many cases, could be overcome by applying artificial intelligence (AI). Therefore, sixth–generation (6G) wireless technologies are being developed to address the limitations of 5G networks. Traditional machine learning (ML) techniques are driven in a centralized way. However, the huge volume of produced wireless data, the confidentiality concerns, and the growing computing competencies of wireless edge devices have led to the exposure of a promising solution in a decentralized way which is called distributed learning. This paper provides a comprehensive analysis of distributed learning (e.g., federated learning (FL), multi–agent reinforcement learning (MARL)–based FL framework) and how to deploy in an effective and efficient way for wireless networks. Moreover, we describe a timely comprehensive review of the role of FL in facilitating 6G enabling technologies, such as mobile edge computing, network slicing, satellite communications, terahertz links, blockchain, and semantic communications. Also, we identify and discuss several open research issues related to FL–empowered 6G wireless networks. In particular, we focus on FL for enabling an extensive range of smart services and applications. For each application, the main motivation for using FL along with the associated challenges and detailed examples for use scenarios are given. Regarding the AI techniques, we consider MARL–based FL framework tailored to the needs of future wireless networks for ensuring fast convergence and high model accuracy of large state and action spaces. Particularly, to manage the fast varying radio channels and limited radio resources (e.g., transmission power and radio spectrum) in a cellular communication environment, this article proposes a robust MARL–based FL framework to enable local users to perform distributed power allocation, mode selection, resource allocation, and interference management. Finally, the paper outlines several prospective upcoming research topics, aimed to create constructive incorporation of MARL–based FL framework for future wireless networks.</p>

show abstract

“…Applying deep learning (DL) techniques to satellite systems has been studied from various aspects, e.g., predicting optimal decoding order for NOMA-based satellite systems [13], assisting beam-timeslot scheduling for beam hopping satellite systems [14], compensating for the non-linear distortion at receivers in LEO satellite systems [15], etc. Applying DL to address constrained combinatorial problems is, however, studied to a limited extent in the literature.…”

Section: Introductionmentioning

confidence: 99%

Dual-DNN Assisted Optimization for Efficient Resource Scheduling in NOMA-Enabled Satellite Systems

Wang

Lei

Lagunas

et al. 2021

2021 IEEE Global Communications Conference (GLOBECOM)

View full text Add to dashboard Cite

In this paper, we apply non-orthogonal multiple access (NOMA) in satellite systems to assist data transmission for services with latency constraints. We investigate a problem to minimize the transmission time by jointly optimizing power allocation and terminal-timeslot assignment for accomplishing a transmission task in NOMA-enabled satellite systems. The problem appears non-linear/non-convex with integer variables and can be equivalently reformulated in the format of mixedinteger convex programming (MICP). Conventional iterative methods may apply but at the expenses of high computational complexity in approaching the optimum or near-optimum. We propose a combined learning and optimization scheme to tackle the problem, where the primal MICP is decomposed into two learning-suited classification tasks and a power allocation problem. In the proposed scheme, the first learning task is to predict the integer variables while the second task is to guarantee the feasibility of the solutions. Numerical results show that the proposed algorithm outperforms benchmarks in terms of average computational time, transmission time performance, and feasibility guarantee.

show abstract

Deep Neural Network-Based Receiver for Next-Generation LEO Satellite Communications

Cited by 7 publications

References 16 publications

Spiking Neural Networks for Detecting Satellite-Based Internet-of-Things Signals

Spiking Neural Networks for Detecting Satellite-Based Internet-of-Things Signals

Distributed Learning for 6G–IoT Networks: A Comprehensive Survey

Dual-DNN Assisted Optimization for Efficient Resource Scheduling in NOMA-Enabled Satellite Systems

Contact Info

Product

Resources

About